| Legendre正交基前向神经网络的权值直接确定法 |
| |
| 引用本文： | 张雨浓,刘巍,易称福,李巍.Legendre正交基前向神经网络的权值直接确定法[J].大连海事大学学报(自然科学版),2008,34(1):32-36. |
| |
| 作者姓名： | 张雨浓 刘巍 易称福 李巍 |
| |
| 作者单位： | 中山大学电子与通信工程系 广州510275(张雨浓，易称福，李巍)，中山大学软件学院 广州510275(刘巍) |
| |
| 基金项目： | 国家自然科学基金资助项目(60643004) |
| |
| 摘 要： | 为避免权值反复迭代修正的冗长BP训练过程,避免传统方法陷入局部极小点,根据多项式理论,构造了一种新型前向神经网络模型,推导了基于最速下降法的误差反传算法和基于伪逆的直接确定法.仿真结果显示,迭代方法和伪逆直接确定法都能达到比较高的工作精度(10-6).
|
| 关 键 词： | 正交多项式 Legendre正交基 标准BP算法 伪逆 |
| 文章编号： | 1006-7736（2008）01-0032-05 |
| 收稿时间： | 2007-10-30 |
| 修稿时间： | 2007年10月30 |
Weights immediate determination for feed-forward neural network with Legendre orthogonal basis function |
| |
| ZHANG Yu-nong,LIU Wei,YI Cheng-fu,LI Wei.Weights immediate determination for feed-forward neural network with Legendre orthogonal basis function[J].Journal of Dalian Maritime University,2008,34(1):32-36. |
| |
| Authors: | ZHANG Yu-nong LIU Wei YI Cheng-fu LI Wei |
| |
| Abstract: | To avoid lengthy iterative BP-training and its local-minima problem,a feed-forward neural network was developed using Legendre orthogonal polynomial activation function based on polynomial interpolation and curve-fitting theory.The neural weights-updating formula was derived by adopting the standard BP training method,and then a pseudo-inverse based method was proposed.Simulation results show that one-step weights-determination method could efficiently achieve at least equally small computational error(10-6) compared with iterative-training method. |
| |
| Keywords: | orthogonal polynomials Legendre orthogonal basis standard back propagation(BP)algorithm pseudo-inverse |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
